Skip to content

[ baobao ] feat: Multi-Dimensional Quality Scoring for Structured Outputs ($10 bounty)#19

Open
mkcash wants to merge 2 commits into
Mint-Claw:mainfrom
mkcash:bounty/1-quality-scoring
Open

[ baobao ] feat: Multi-Dimensional Quality Scoring for Structured Outputs ($10 bounty)#19
mkcash wants to merge 2 commits into
Mint-Claw:mainfrom
mkcash:bounty/1-quality-scoring

Conversation

@mkcash
Copy link
Copy Markdown

@mkcash mkcash commented May 30, 2026

Implementation

Built a multi-dimensional quality scoring engine for structured outputs (JSON, markdown, code, text).

Features

  • Auto-detects output format (JSON, Markdown, Code Block, Short Text, etc.)
  • Scores all 5 dimensions: Completeness (0.30), Format Compliance (0.20), Coverage (0.25), Clarity (0.15), Validity (0.10)
  • Returns weighted score (0-1), quality rating, per-dimension scores, and feedback
  • Pass threshold check (>=0.50)
  • Batch scoring: 100 submissions in ~0.02s (well under the 10s requirement)

Usage

python3 quality_scorer.py          # Interactive mode
python3 quality_scorer.py --test   # Run tests
cat subs.txt | python3 quality_scorer.py --batch  # Batch mode

Payment

Closes #1

mkcash and others added 2 commits May 29, 2026 15:50
Implements a scoring algorithm for structured submissions (JSON, Markdown, Code, Text)
against a rubric, returning 0-1 weighted scores with per-dimension feedback.

- Auto-detects submission format
- Scores 5 dimensions: Completeness, Format Compliance, Coverage, Clarity, Validity
- Custom rubric support
- 12 unit tests all passing
- 100 submissions processed in <105ms (requirement: <10s)
- NLP feedback generation
- Sample scorecards included
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[BOUNTY $10] Multi-Dimensional Quality Scoring for Structured Outputs

1 participant